30 Day Mobile App Programs for Stress Management in Customer Service Representatives

A Randomized Controlled Trial Comparing a 30 Day Mediation App to 30 Day Active Problem Solving App: Stress Management in Customer Service Representatives

Burnout and stress at work can make individuals less productive, which can carry over into their personal and at-home lives and negatively impact health. Customer service representatives are under especially high strain as they are exposed to significant interpersonal conflict at work, both with frustrated customers and with pressure from coworkers and supervisors. However, recent research has found that different stress management interventions (e.g., mindfulness meditation training) can increase job satisfaction and work productivity. However, individuals with significant stress might find these training program classes difficult to attend with their busy schedules. Newer interventions have focused on smartphone mobile applications as an effective delivery system for these training programs. Thus, the purpose of this project is to conduct a randomized controlled trial (RCT) comparing two different stress management smartphone app programs to evaluate effects on job-related outcomes, functional and structural brain outcomes, and biology.

Approach: Employees will be given study flyers describing the study and encouraged to call a project hotline to evaluate their study eligibility. Eligible participants will be scheduled for an in-person study baseline and neuroimaging appointment at Carnegie Mellon University where they will provide written informed consent, complete baseline survey measures, provide several drops of blood via a finger prick (Dried Blood Spot (DBS) samples), and complete a 60-minute neuroimaging scan. After completing baseline measures participants will be instructed on how to download and use the Headspace smartphone app, and will be randomly assigned to one of the two programs. Participants will be asked to complete a 7- minute end-of-day daily diary measure each day during the treatment period, which will measure stress, affect, sleep, and workplace perceptions (a text message link will be sent an hour before the participant's standard bedtime each day). The project team will monitor treatment adherence (participants who do not complete a daily practice session will be called and reminded the following day). In the week following the end of the treatment period, participants will be scheduled to come back to Carnegie Mellon for a post-treatment appointment where they will complete the same measures as at baseline (survey measures, DBS, and neuroimaging). Participants will then be debriefed, compensated, and dismissed. Approximately two months after participants complete the 30-day program, we will call them for a 15-minute follow-up call. During this call, participants will answer a few questionnaire items and a brief program evaluation. Interventions: Participants will be asked to complete one lesson daily for 30- days on the Headspace app; the first ten lessons will be 10 minutes long, the next ten will be 15 minutes long, and the final ten will be 20 minutes long. The mindfulness program will consist of the standard base program offered in Headspace. The Recharge program will consist of problem solving and stress management instruction. Setting and Participants: David Creswell's Health and Human Performance Laboratory at Carnegie Mellon University will direct this study. Participants will be customer service employees (those who interact with customers daily via phone or in-person interactions) recruited from the Pittsburgh region.

The outcomes assessor will remain masked to which program the participant will use. However the experimenter will be helping set up the participants with the app content and thus will not be masked to condition.

30-day smartphone based mindfulness training intervention consisting of 10-minutes for the first 10 days, 15 minutes for the next 10 days, and 20 minutes for the final 10 days.

30-day smartphone based reflection and problem solving training intervention consisting of 10-minutes for the first 10 days, 15 minutes for the next 10 days, and 20 minutes for the final 10 days.

Evaluated using a burnout scale for customer service representatives. This scale is used to measures feelings of burnout, a psychological disposition characterized by emotional exhaustion, lack of personal accomplishment, and a tendency to depersonalize others. The 24-item scale uses a 6-point Likert scale, and items are summed to create a composite score, with a range of 24-144, higher scores indicating greater burnout. Analyses will focus on a time (baseline, post-intervention, follow-up) by group interaction, and subsequent within group and within timepoint main effects. Here we hypothesize that while there will be a main effect of time, there will be a significant time by group interaction on reducing burnout.

Evaluated using Job Content Questionnaire. The Job Content Questionnaire is designed to measure the social and psychological characteristics of the workplace environment such as psychological demands, decision latitude, social support, physical demands, and job insecurity. The Job Stress measure will be calculated by combining the Psychological Demands and Decision Latitude subscales. The Psychological Demands subscale is 5-items and the Decision Latitude is 9-items. All are scored on a 1-4 Likert, resulting in a range of 14-56, scored such that higher numbers indicate greater job stress. Analyses will focus on a time (baseline, post-intervention, follow-up) by group interaction, and subsequent within group and within timepoint main effects. Here we hypothesize that while there will be a main effect of time, there will be a significant time by group interaction on reducing job stress.

Evaluated using Job Satisfaction Questionnaire. This is a 10-item scale, each item scored on a 1-5 Likert scale, resulting in a range of 10-50, with higher scores indicating greater job satisfaction as follows: 10-26: very low satisfaction 27-31: low satisfaction 32-38: average satisfaction 39-41: high satisfaction 42-50: very high satisfaction Analyses will focus on a time (baseline, post-intervention, follow-up) by group interaction, and subsequent within group and within timepoint main effects. Here we hypothesize that while there will be a main effect of time, there will be a significant time by group interaction on increasing job satisfaction.

Evaluated using Work Life Conflict Scale.The Work Life Conflict designed to measure conflict between work and non-work experiences. The 5-item measure has been used in previous work, and participants respond to each item using a 1-7 Likert scale with higher numbers indicating greater work-life conflict. Some previous work has averaged responses to the 5-items, thus resulting in a range of 1-7 for this scale. Analyses will focus on a time by group interaction, and subsequent within group and within timepoint main effects. Here we hypothesize that while there will be a main effect of time, there will be a significant time by group interaction on reducing work-life conflict.

Evaluated using Life Work Conflict Scales.The Life Work Conflict designed to measure conflict between non-work experiences and the individual's experiences at work. The 5-item measure has been used in previous work, and participants respond to each item using a 1-7 Likert scale with higher numbers indicating greater life-work conflict. Some previous work has averaged responses to the 5-items, thus resulting in a range of 1-7 for this scale. Analyses will focus on a time by group interaction, and subsequent within group and within timepoint main effects. Here we hypothesize that while there will be a main effect of time, there will be a significant time by group interaction on reducing life-work conflict.

Evaluated by asking at 2-month post-intervention follow-up. Analyses will focus on a main effect of group on lower number of sick days taken.

Functional neural activity to a task displaying positive vs. neutral images. Analyses will focus on a time (baseline, post-intervention) by group (Headspace, Recharge) interaction, and subsequent within group and within timepoint main effects. ROI and connectivity analyses will focus on the reward network. Correlations with changes in included questionnaires and biology will also be assessed.

Functional neural activity to a task displaying negative vs. neutral images. Analyses will focus on a time (baseline, post-intervention) by group (Headspace, Recharge) interaction, and subsequent within group and within timepoint main effects. ROI and connectivity analyses will focus on the threat/salience network. Correlations with changes in included questionnaires and biology will also be assessed.

Resting state functional connectivity will also be assessed using a time by group interaction and time and group main effects. We will conduct analyses focusing on the default mode and frontostriatal networks. Correlations with changes in included questionnaires and biology will also be assessed.

Diffusion spectrum analyses will include time x group changes in white matter using fractional anisotropy (FA). Analyses will focus on whole brain FA (i.e., white matter integrity across the whole brain). Further analyses will explore associations between changes in white matter and our other outcomes of interest (behavior, biology, affect, functional task performance, etc.).

Structural changes in the amygdala will also be assessed using time by group interaction and time and group main effects. Voxel based morphometry techniques will be used. In particular, change in gray matter volume will also be correlated with our relevant outcomes to assess whether changes in structure are associated with important changes in behavior, biology or affect.

Structural changes in the reward network will also be assessed using time by group interaction and time and group main effects. Voxel based morphometry techniques will be used. In particular, these changes in gray matter volume will also be correlated with our relevant outcomes to assess whether changes in structure are associated with important changes in behavior, biology or affect.

Changes in gene expression will be assessed using a time by group interaction (with time and group main effects also examined as ancillary findings, and in the event a significant time by group interaction is observed, follow-up group-stratified "simple effects" analyses of time effects may also be conducted to help interpret results). Analyses will focus an a priori-specified gene regulation pattern involving increased expression of inflammation-related genes and decreased expression of antiviral gene - a pattern called the conserved transcriptional response to adversity (CTRA).

Measures will be assessed end-of-day ("I was productive at work today." on 1-7 agreement Likert scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased productivity.

Measures will be assessed end-of-day ("I feel like I did something meaningful at work today." on 1-7 agreement Likert scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased self-reports of meaningful activities at work.

Measures will be assessed end-of-day ("At work today, I found myself doing things without really paying attention." adapted from Mindful Attention and Awareness Scale, answered on 1-6 frequency scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased attention at work.

Interpersonal conflict with customers and coworkers will be assessed by the following questions (1-"none at all" to 5-"a great deal" scale): How much conflict did you experience with customers today? How much conflict with coworkers and managers/supervisors did you experience today? These items were selected to capture the listed outcome measure and do not come from a previously used scale. The outcome measure will be assessed end-of-day during the 30 day period. Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on reduced interpersonal conflict with customers and coworkers.

Measures will be assessed end-of-day using three items from Perceived Stress Scale adapted for work. Answered on a 0-never to 4-very often scale, with higher scores indicating greater perceived stress at work. Analyses will use multilevel modeling to explore differences between groups. We hypothesize a main effect of time, and a possible time by group interaction effect on perceived stress.

Evaluated using Perceived Stress Scale (PSS). The PSS measures the degree to which situations in one's life are appraised as stressful. Individual scores on the PSS can range from 0 to 40 with higher scores indicating higher perceived stress. Scores ranging from 0-13 would be considered low stress. Scores ranging from 14-26 would be considered moderate stress. Scores ranging from 27-40 would be considered high perceived stress. Here, we expect a main effect of time, with a possible time by group interaction on lowered perceived stress.

Evaluated using the SF-12, a 12-item scale assessing physical health and mental health. Composite scores range from 0-100 with higher scores indicating better health. Here, we expect a main effect of time, with a possible time by group interaction on improved self-reported physical health.

Evaluated using the CES-D which measures depressive symptoms over the last two weeks using 20-items. The total score is a sum of the items and scores can range from 0-60, with higher scores indicating greater depressive symptoms. A score equal to or above 16 can indicate a person at risk for clinical depression. Here, we expect a main effect of time, with a possible time by group interaction on lowered depressive symptoms

Evaluated using the Five Factor Mindfulness Questionnaire (FFMQ). This FFMQ is based on a factor analytic study of five independently developed mindfulness questionnaires. The analysis yielded five factors that appear to represent elements of mindfulness as it is currently conceptualized. The five facets are observing, describing, acting with awareness, non-judging of inner experience, and non-reactivity to inner experience. The 39-item scale is scored on 1-very rarely true to 5-very often true scale, range 39-195, with higher scores indicating greater mindfulness. Here, we expect a time by group interaction on increased mindfulness.

Evaluated using the Social Provisions Scale.The Social Provisions Scale was developed to assess the provisions of social relationships and what we receive from relationships with other people. The six provisions include guidance (advice or information), reliable alliance (assurance that others can be counted on in times of stress), reassurance of worth (recognition of one's competence), attachment (emotional closeness), social integration (a sense of belonging to a group of friends), and opportunity for nurturance (providing assistance to others) across 24 items. These subscales are summed to create a score for each subscale (range of 4-16), and subscales are summed together to create a composite total score (range of 24-96) with higher scores indicating higher perceptions of social support. Here, we will explore a possible time by group interaction on increased perceptions of available social support.

Evaluated using the UCLA Loneliness Scale. The UCLA Loneliness scale is designed to measure one's subjective feelings of loneliness as well as feelings of social isolation. Scores are summed from each item and higher scores indicate greater degrees of loneliness (range 20-80). Here, we will explore a possible time by group interaction on reduced loneliness.

Evaluated using the Distress Tolerance Scale which assesses the perceived ability to tolerate distress using 16-items, summed, with a range of 16-90 with higher scores indicating greater levels of distress tolerance. Here, we will explore a possible time by group interaction on increased acceptance.

Evaluated using the White Bear Thought Suppression Inventory. 15 items answered on a 5-point Likert scale and summed, resulting in a range of 15-75 (higher scores indicating greater thought suppression). Greater thought suppression is believed to be an indicator of low equanimity. Here, we will explore a possible time by group interaction on increased equanimity (reduced thought suppression scores).

CRP levels will be assessed through dried blood spots collected at baseline and post-intervention. Changes in CRP levels will be assessed using a time by group interaction and time and group main effects. We hypothesize a main effect of time, with exploratory analyses to examine the interaction effect.

Measures will be assessed end-of-day ("How positive was your mood today?" on 1-not at all to 7-extremely scale). Analyses will use multilevel modeling to explore differences between groups. We hypothesize a time by group interaction effect on positive affect.

Measures will be assessed end-of-day ("How positive was your mood today?" on 1-not at all to 7-extremely scale). Analyses will use multilevel modeling to explore differences between groups. We hypothesize a main effect of time on negative affect.

Measures will be assessed end-of-day ("Rate how you feel about today's shift at work." on a 1-extremely bad to 7-extremely good scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on more positive feelings about work.

Measures will be assessed end-of-day ("I had a positive attitude at work today." on a 1-7 agreement Likert scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased positive attitude at work.

Measures will be assessed end-of-day using one item from the Philadelphia Mindfulness Scale ("There were things I tried not to think about while I was at work today." reverse scored such that higher scores reflect more acceptance). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased acceptance.

Measures will be assessed end-of-day ("The demands of my work interfered with my life away from work today." on a 1-7 agreement Likert scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on reduced work-life conflict.

Measures will be assessed end-of-day ("Personal life issues interfered with my ability to perform work-related duties today." on a 1-7 agreement Likert scale). Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on reduced life-work conflict.

Measures will be assessed end-of-day. Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on a higher number of social interactions outside of work.

Measures will be assessed end-of-day and ask the participant to reflect on how connected they felt that day. Analyses will use multilevel modeling to explore differences between groups. Here, we will explore a possible time by group interaction on increased feelings of social connection.

Measures will be assessed end-of-day. Analyses will use multilevel modeling to explore differences between groups. We hypothesize a main effect of time, and a possible time by group interaction effect on increased hours asleep.

Inclusion Criteria: Fluent in English 18-70 years old Planning to remain in the Pittsburgh area for the duration of participation Endorse significant work stress Exclusion Criteria: Any significant mental and physical health conditions Metal implants Significant claustrophobia

The research team will share data associated with self-reports (e.g. job satisfaction, stress levels) and gene expression assays. All data and documentation will be de-identified and will be consistent with applicable laws and regulations. Neuroimaging data will be defaced such that no identifiable information remains on the images.When publishers require analyzed data to be available in order to publish manuscripts, those data will be provided by link to the repository. Data will be shared via KiltHub, a comprehensive repository hosted by Carnegie Mellon University. Data will be shared when a publisher requests that information be shared as part of the terms of publication. A file containing all variables necessary to replicate the analyses from the manuscript will be uploaded. If neuroimaging data is to be shared, the relevant scan will be shared after being de-identified, in addition to the necessary structural scan (as it is required for preprocessing co-registration).

Study data will become available when journals ask for these items upon publication of a paper from this dataset.

Data relevant to replicating analyses will be shared via KiltHub, a comprehensive repository hosted by Carnegie Mellon University. This repository helps keep track of who has accessed data, and is supported by the university and thus be available to other researchers.

Slutsky J, Chin B, Raye J, Creswell JD. Mindfulness training improves employee well-being: A randomized controlled trial. J Occup Health Psychol. 2019 Feb;24(1):139-149. doi: 10.1037/ocp0000132. Epub 2018 Oct 18.

Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983 Dec;24(4):385-96. No abstract available.

Russell D, Peplau LA, Cutrona CE. The revised UCLA Loneliness Scale: concurrent and discriminant validity evidence. J Pers Soc Psychol. 1980 Sep;39(3):472-80. doi: 10.1037//0022-3514.39.3.472.

Wegner DM, Zanakos S. Chronic thought suppression. J Pers. 1994 Dec;62(4):616-40. doi: 10.1111/j.1467-6494.1994.tb00311.x.

Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994). Behavioral and psychological consequences of boundary spanning burnout for customer service representatives. Journal of Marketing Research, 558-569.

Schnall PL, Schwartz JE, Landsbergis PA, Warren K, Pickering TG. A longitudinal study of job strain and ambulatory blood pressure: results from a three-year follow-up. Psychosom Med. 1998 Nov-Dec;60(6):697-706. doi: 10.1097/00006842-199811000-00007.

Macdonald, S., & Maclntyre, P. (1997). The generic job satisfaction scale: Scale development and its correlates. Employee Assistance Quarterly, 13(2), 1-16.

Waumsley, J. A., Houston, D. M., & Marks, G. (2010). What about us? Measuring the work- life balance of people who do not have children. Review of European Studies , 2(2), 3- 17.

Cole SW, Conti G, Arevalo JM, Ruggiero AM, Heckman JJ, Suomi SJ. Transcriptional modulation of the developing immune system by early life social adversity. Proc Natl Acad Sci U S A. 2012 Dec 11;109(50):20578-83. doi: 10.1073/pnas.1218253109. Epub 2012 Nov 26.

Brown KW, Ryan RM. The benefits of being present: mindfulness and its role in psychological well-being. J Pers Soc Psychol. 2003 Apr;84(4):822-48. doi: 10.1037/0022-3514.84.4.822.

Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996 Mar;34(3):220-33. doi: 10.1097/00005650-199603000-00003.

Radloff, L.S. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1977.

Baer RA, Smith GT, Hopkins J, Krietemeyer J, Toney L. Using self-report assessment methods to explore facets of mindfulness. Assessment. 2006 Mar;13(1):27-45. doi: 10.1177/1073191105283504.

Cutrona, C. E., & Russell, D. W. (1987). The provisions of social relationships and adaptation to stress. Advances in personal relationships, 1(1), 37-67.

Simons, J. S., & Gaher, R. M. (2005). The Distress Tolerance Scale: Development and validation of a self-report measure. Motivation and Emotion, 29(2), 83-102.

Cardaciotto L, Herbert JD, Forman EM, Moitra E, Farrow V. The assessment of present-moment awareness and acceptance: the Philadelphia Mindfulness Scale. Assessment. 2008 Jun;15(2):204-23. doi: 10.1177/1073191107311467. Epub 2008 Jan 9.